To accelerate the application of lead-free inorganic halide perovskites in solar cells, it is necessary to develop novel perovskite materials with appropriate bandgap widths, high stability, and environmental friendliness. This represents a crucial pathway for driving innovation in photovoltaic technology and reducing reliance on traditional fossil fuels. However, traditional material development paradigms heavily rely on trial-and-error experimental screening or pure density functional theory (DFT) calculations, which results in significant time and material costs.

To address these challenges, this study innovatively proposes and implements an efficient screening strategy that leverages the synergy between deep learning and DFT calculations. By constructing a database containing 1181 inorganic halide double perovskite materials, we systematically train and compare the performance of five mainstream machine learning models for the bandgap prediction task: random forest regression (RFR), gradient boosting regression (GBR), support vector regression (SVR), extreme gradient boosting regression (XGBR), and a deep neural network (DNN) model. The results demonstrate that the DNN model, by using its powerful nonlinear mapping capability and advantage in automatic high-dimensional feature extraction, achieves exceptional prediction accuracy on the test set, with the mean absolute error (MAE) is significantly reduced to 0.264 eV, and the coefficient of determination (R²) reaches 0.925. Its performance is markedly superior to other compared models, highlighting the immense potential of deep learning in predicting complex material properties.

Using this optimized DNN model, this study successfully screens four promising inorganic double perovskite candidates from 55 potential materials: Cs₂GaAgCl₆, Cs₂AgIrF₆, Cs₂InAgCl₆, and Cs₂AlAgBr₆. Among them, Cs₂AgIrF₆ and Cs₂AlAgBr₆ perform particularly well, with predicted bandgaps of 1.36 eV and 1.20 eV, respectively. This range ideally matches the requirement for efficient light absorption in solar cells. Further device performance simulations reveal that the solar cell based on Cs₂AgIrF₆ achieves a simulated power conversion efficiency (PCE) of 23.71%, with an open-circuit voltage (V_OC) of 0.94 V, a short-circuit current density (J_SC) of 31.19 mA/cm², and a fill factor (FF) of 80.81%. Cs₂AlAgBr₆ also exhibits a simulated efficiency of 22.37%, corresponding to V_OC = 0.78 V, J_SC = 36.73 mA/cm², and FF = 77.66%. Notably, both materials exhibit high open-circuit voltages and fill factors, clearly demonstrating excellent carrier separation efficiency and significantly reducing non-radiative recombination losses within these materials.

In summary, this study successfully validates the significant efficacy of the deep learning-DFT synergistic strategy in accelerating the discovery of novel lead-free perovskite materials. This method not only greatly improves the efficiency of DFT data analysis and the depth of pattern mining, overcoming some bottlenecks related to traditional high-throughput calculations, but more importantly, it provides a practical and highly innovative approach for rationally designing high-performance, stable, and environmentally friendly lead-free perovskite solar cells, which holds positive significance for promoting green, low-carbon energy technologies.